Writing our own K Nearest Neighbors in Code - Practical Machine Learning Tutorial with Python p.17

TL;DR

This content discusses the implementation of a K nearest neighbors function, including the calculation of Euclidean distance. It also mentions the use of numpy and compares the function to scikit-learn's K nearest neighbors algorithm.

Transcript

what is going on everybody and welcome to part 17 of our machine learning tutorial series we've been working on K nearest neighbors and let's get into it don't want to waste any more time here so we've started creating this K nearest neighbors function all we've done so far though is just warn the user when they're trying to do something stupid so ... Read More

Key Insights

• 👈 Creating a K nearest neighbors function involves calculating Euclidean distances between a prediction point and other data points.
• 💨 Numpy arrays can be used to perform a faster version of Euclidean distance calculation.
• 😥 The function appends distances to a list and sorts it to find the K nearest data points.
• 👥 The most common group is determined using the Counter object from the collections module.

Questions & Answers

Q: What is the purpose of the K nearest neighbors function?

The K nearest neighbors function is used to calculate the K nearest data points to a prediction point based on the comparison of Euclidean distances.

Q: What is the problem with K nearest neighbors when comparing data points?

The problem is that K nearest neighbors requires comparing the prediction point to all other points in the dataset, making it a slow calculation.

Q: How can the Euclidean distance calculation be improved?

Numpy arrays can be used to perform a faster calculation of Euclidean distance, with a simplified version also available. This saves time and allows for calculations in higher dimensions.

Q: How is the most common group determined in the K nearest neighbors function?

The K nearest neighbors function uses the Counter object from Python's collections module to determine the most common group, taking into account the K nearest data points.

Summary & Key Takeaways

• The content focuses on creating a K nearest neighbors function and discusses the problem of comparing a data point to other data points to find the closest ones.

• It explains different versions of the Euclidean distance calculation, including a faster method using numpy arrays.

• The content also covers appending distances to a list, sorting the list, and obtaining the most common group.

• Lastly, it mentions the intention to compare the function to scikit-learn's K nearest neighbors algorithm in the next tutorial.